Machine Learning for COVID-19 Patient Management: Predictive Analytics and Decision Support

Publication date: Feb 27, 2024

Background. The global impact of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has profoundly affected economies and healthcare systems around the world, including Lebanon. While numerous meta-analyses have explored the systemic manifestations of COVID-19, few have linked them to patient history. Our study aims to fill this gap by using cluster analysis to identify distinct clinical patterns among patients, which could aid prognosis and guide tailored treatments. Methods. We conducted a retrospective cohort study at Beirut’s largest teaching hospital on 556 patients with SARS-CoV-2. We performed cluster analyses using K-prototypes, KAMILA and LCM algorithms based on 26 variables, including laboratory results, demographics and imaging findings. Silhouette scores, concordance index and signature variables helped determine the optimal number of clusters. Subsequent comparisons and regression analyses assessed survival rates and treatment efficacy according to clusters. Results. Our analysis revealed three distinct clusters: “resilient recoverees” with varying disease severity and low mortality rates, “vulnerable veterans” with severe to critical disease and high mortality rates, and “paradoxical patients” with a late presentation but eventual recovery. Conclusions. These clusters offer insights for prognosis and treatment selection. Future studies should include vaccination data and various COVID-19 strains for a comprehensive understanding of the disease’s dynamics.

Concepts Keywords
Austria Certified
Cat Cluster
Immunocompromised Clusters
Youngest Covid


Type Source Name
disease MESH COVID-19
disease VO Severe acute respiratory syndrome coronavirus 2
disease IDO history
disease VO Gap
drug DRUGBANK Lincomycin
disease VO vaccination
disease MESH severe acute respiratory syndrome
disease MESH sore throat
disease MESH respiratory failure
disease MESH fatal outcome
disease MESH metabolic diseases
disease MESH nosocomial infections
disease MESH death
disease IDO algorithm
drug DRUGBANK Creatinine
disease MESH Chronic renal failure
disease MESH vascular disease
disease MESH heart failure
disease MESH Diabetes mellitus
drug DRUGBANK Dextrose unspecified form
disease MESH Hypertension
disease MESH stroke
disease MESH cardiac events
disease MESH cancer
disease MESH autoimmunity
disease MESH COPD
disease MESH asthma
pathway KEGG Asthma
disease MESH pulmonary fibrosis
disease MESH inflammation
disease MESH infection
disease VO volume
drug DRUGBANK Oxygen
disease IDO blood
disease MESH clinical relevance
disease IDO process
disease IDO symptom
drug DRUGBANK Aspartame
disease VO dead
disease MESH Pulmonary Diseases
disease VO population
disease MESH complications
drug DRUGBANK Tocilizumab
drug DRUGBANK Azithromycin
drug DRUGBANK Hydroxychloroquine
drug DRUGBANK Acetylsalicylic acid
disease MESH superinfection
disease MESH bleeding
drug DRUGBANK Doxycycline
drug DRUGBANK Prednisone
drug DRUGBANK Baricitinib
disease MESH critical illness
disease MESH pneumonia
drug DRUGBANK Medical air

(Visited 1 times, 1 visits today)